Research on Multilevel Classification of High-Speed Railway Signal Equipment Fault Based on Text Mining

Table 2

Training results of BiGRU and BiLSTM neural networks.

Method	Level	Accuracy	Recall	F1 value
BiGRU	First-level classification	0.8745	0.8579	0.8661
	Second-level classification	0.7456	0.6809	0.7117
BiLSTM	First-level classification	0.8671	0.8368	0.8517
	Second-level classification	0.6713	0.6025	0.6350